SPICILEGIA FLOE.E SINENSIS. 233 



57. Aristolochia Kcem2}feri,Willd. In. Silver Island, Chin-kiang, 

 Maio, 1863, coll. Hay. I have seen no Japanese specimens, and 

 mine is not in good condition, but it seems to agree very well with 

 Duchartre's diagnosis. It has not hitherto been recorded as a 

 native of China. I have another Aristolochia from the woods of 

 the Tsing-yiin pass, prohably allied to A. mclica, Linn., but the 

 leaves, borne on petioles three inches long, are more like those of 

 a Dioscorea, ovate-oblong in contour, 6-7 inches long, 4 inches 

 broad, with a wide sinus at the base, and rounded auricles 1^ inch 

 in length, produced downwards. The capsule is ovoid for 1^ inch, 

 and drawn out at the base into an equally long stipes splitting into 

 as many divisions as the fruit. There are no iiowers, and I cannot 

 venture to characterise the i)lant. 



58. Hahenaria sagittifera, Ecbb. fil. Circa Chi-fu, 1873, coll. 

 C. C. Stuhlmann ; prope Ta-chiao-sz, Chinae bor., 1874, unicum 

 specimen invenit Dr. 0. F. a Moellendorff. Only known previously 

 from Japan and Manchuria. 



59 . Pollia japonica, Thunb . In silvis supra coenobium buddhicum 

 Fi-loi-tsz, ad fauces fl. North Eiver Tsing-jain dictas, m. Jul. 1865, 

 detexit Samj)soi\; Sept., 1866, coll. Sami)son et Hance. 



60. Carex Fioyleana, N. ab E. Cu'ca Amoy, Martio, 1863, coU. 

 C. F. M. de Grijs. Reduced latterly by Dr. Boott, who had 

 formerly considered it distinct, to C. breviculmis, R. Br., from which 

 it differs by the culms shorter in proportion to the leaves, the 

 female spikes usually male at the summit, and the squamae ovate 

 and gradually attenuated into — not broadly shouldered and truncate 

 at the base of — the cusps, which are much shorter, so that the 

 s^Dikes have not the aristate apx^earance of the other. The Chinese 

 specimens agree in aU respects with the beautiful plate (* 111. 

 Carex,' i., t. 19.) 



61. Carex tristacJuja, Thunb. In rupibus humidis insulae 

 Danorum, Whampoae, primo inveni, A]}y., 1861; in montibus 

 Pak-wan, supra Cantonem, raram vidit Sampson. The late Dr. 

 Boott determined this as his C. monadelpha, but he subsequently 

 informed me that he considered me right in regarding the latter as 

 identical with Thunberg's species. Schkuhr's figure ('Riedgr. 

 Ww.,' 109) though representing a fragment only, is not uncharac- 

 teristic, but the male spike is erroneously coloured quite blackish. 



62. Panicum humile, N. ab E. In agris requietis, Whampo^, 

 ipse primum legi, m. Sept., 1863. The Chinese specimens are 

 precisely similar to those from Ceylon, whence the species had, I 

 believe, only been previously recorded. 



63. Pollinia imherbis, N. ab E. In udis secus viam ad montes 

 Pak-wan, extra Cantonem, ducentem, m. Jan., 1868, detexit 

 Sampson. 



In aspect rather resembhng a dwarfed Leersia than an Andro- 

 pogonea. 



64. Ischcpmum? segetum, Trin. Secus fl. Lien-chau, prov. Can- 

 tonensis, Octobri, 1876, coll. Rev. J. C. Nevin. I refer the Chinese 

 grass to this species merely from its agreement with Trinius' brief 

 character ('Mem. Acad. St. Petersb.,' 6e. ser., ii., 294.) It is of a 



2h 



